Update app.py

app.py

@@ -11,7 +11,8 @@ st.title("🩺 ChatDoctor - Medical Assistant")
 # Load model and tokenizer
 @st.cache_resource
 def load_model():
-    model = AutoModelForCausalLM.from_pretrained("abhiyanta/chatDoctor", use_cache=True).to("cpu")
+    # Ensure no GPU (CUDA) optimizations like quantization
+    model = AutoModelForCausalLM.from_pretrained("abhiyanta/chatDoctor", use_cache=True)
     tokenizer = AutoTokenizer.from_pretrained("abhiyanta/chatDoctor")
     return model, tokenizer
 
@@ -33,7 +34,7 @@ if st.button("Ask ChatDoctor"):
             ""
         )
 
-        # Tokenize and move to CPU
+        # Tokenize and ensure it's on CPU
         inputs = tokenizer([formatted_prompt], return_tensors="pt").to("cpu")
 
         # Stream the generated output
@@ -48,4 +49,4 @@ if st.button("Ask ChatDoctor"):
 
 # Footer
 st.markdown("---")
-st.caption("Powered by Hugging Face 🤗 and bitsandbytes ⚡")
+st.caption("Powered by Hugging Face 🤗")